BriefGPT.xyz
Jan, 2019
部署后从对话中学习：聊天机器人，喂饱自己！
Learning from Dialogue after Deployment: Feed Yourself, Chatbot!
HTML
PDF
Braden Hancock, Antoine Bordes, Pierre-Emmanuel Mazare, Jason Weston
TL;DR
本研究提出自我反馈聊天机器人，通过从参与的对话中提取新的训练样本和估计用户满意度来改进聊天机器人的对话能力，并在 PersonaChat chit-chat 数据集上进行实验得到了显著的性能提升。
Abstract
The majority of conversations a
dialogue agent
sees over its lifetime occur after it has already been trained and deployed, leaving a vast store of potential
training signal
untapped. In this work, we propose the
→